Response Differences on the Minecraft Writing Assessment: Comparisons between Children with Autism Spectrum Disorder, Children with Attention-Deficit/Hyperactivity Disorder, and Typically Developing Children

Background: Children with autism spectrum disorder (ASD) demonstrate broad challenges with transcription and text generation. Recent studies have shown that these challenges are related to difficulties with language, social communication, and executive functions. The Minecraft Writing Assessment (MWA) was designed to explore writing challenges in children with ASD compared to children with attention-deficit/hyperactivity disorder (ADHD) or typical development (TD). Preliminary findings have shown group differences on word production but have not explored potential relationships between broader task performance and text production across groups.

Objectives: 1) To examine group differences in response style to the prompt, audience awareness, and items recalled during video recall or writing; 2) to compare relationships between word production with item recall during video recall and writing.

Methods: A total of 132 (66 with HFASD, 29 with ADHD, and 37 TD) children participated who were matched on age (M=14, SD=2.24). Both clinical groups were matched on FIQ. ASD diagnoses were confirmed using the ADOS-2, and ADHD diagnoses were confirmed using the Conners-3. All participants completed the experimenter-designed MWA that included a questionnaire about Minecraft, watching and recalling items from a short two-minute gameplay video, and a ten-minute writing prompt where participants either handwrote or typed a story about the video to a specific audience (a peer their own age who had never played Minecraft). Writing samples were counted for words and were coded as either being written in a narrative or a descriptive style. Audience awareness was coded dichotomously for including at least one comment/detail intended for someone unaware of Minecraft. Item recall was coded for 35 items across seven categories for immediate video recall and inclusion in writing sample (Table 1). All categories were also scored dichotomously for whether a child recalled any item within a given category (Table 2).

Results: Groups did not differ on response style (67-85% wrote descriptively) or audience awareness (28-36% addressed an audience; Table 1). Groups differed on item inclusion, with ASD and ADHD performing lowest on total item inclusion (Table 1). Children with ASD or ADHD often performed lowest for recall during video or writing for items remembered per category (Table 1) and for any items were recalled per category (Table 2). ASD showed a moderate relationship between word count and item inclusion (r=.46, p<.0005) but not video recall (r=.23, p=.07). ADHD showed a moderate relationship with word count and both video recall and item inclusion (rs=.60-.61, ps<.001). Both were unrelated for TD, though prior findings suggested this was due to less variability in observed word count. Partial correlations with FIQ demonstrated similar findings across groups.

Conclusions: Groups responded similarly to the prompt, but children with ASD demonstrated specific difficulties with item recall, particularly with children with ASD showing more prominent gaps in recall categories compared to children with ADHD or TD children. Children with ADHD showed a more consistent relationship compared to children with ASD between recall and inclusion of items with word count. These findings offer insights into potential predictors associated with writing difficulties in children with ASD.